The mmap Device Operation
Memory mapping is one of the most interesting features of modern Unix systems. As far as drivers are concerned, memory mapping can be used to provide user programs with direct access to device memory.
For example, a simple ISA frame grabber holds image data in its own memory, either in the 640KB-1MB address range or in the ``ISA hole,'' the range between 14MB and 16MB (see Section 8.3 in Chapter 8). While copying image data to conventional (and faster) RAM is a suitable approach for casual grabbing, if the user program needs to access the current image frame every now and then, an mmap approach is better suited to the task.
Mapping a device means associating a range of user-space addresses to device memory. Whenever the program reads or writes in the assigned address range, it is actually accessing the device.
As you might suspect, not every device lends itself to the mmap
abstraction; it makes no sense, for instance, for serial ports and
other stream-oriented devices. Another limitation of mmap is
that mapping is
PAGE_SIZE-grained. The kernel can dispose of
virtual addresses only at the level of page tables;
therefore, the mapped area
must be a multiple of
PAGE_SIZE and must live in physical
memory starting at
an address that is a multiple of
PAGE_SIZE. The kernel
accomodates for size-granularity by making a region slightly bigger if
its size isn’t a multiple of the page size. Alignment
issues are usually handled by mucking with
vma->vm_offset, but ...